Hepatitis C virus (HCV) is the principal causative agent for non-A, non-B Hepatitis. The prevalence of HCV infection in the blood donor population has been estimated to range from 0.4 to 2% (Choo et al., 1989). Acute HCV infection leads, in more than 70% of the patients, to the development of chronic hepatitis that can evolve towards cirrhosis and hepatocellular carcinoma (Saito et al., 1990). HCV is an enveloped positive-stranded RNA virus which is classified in the Flaviviridae family (Francki et al., 1991, Miller et al., 1990). It contains a genome of about 9,500 nts encoding a polyprotein of 3010 to 3033 amino acids. Processing of the polyprotein by host and viral proteases results in the production of structural and nonstructural (NS) proteins (Rice et al., 1996). Structural proteins include a nucleocapsid and two putative virion envelope glycoproteins E1 and E2 (Miyamura et al., 1993). Non-structural proteins include NS2 to NS5 antigens.
In some individuals, acute infection successfully resolves indicating that HCV can be controlled by the host immune system. The mechanisms by which the host overcomes HCV infection remain unknown. Previous reports strongly suggest that humans and chimpanzees can generate virus-neutralizing antibodies (Choo et al., 1994, Farci et al., 1994, 1996, Shimizu et al., 1994). Successful in vivo protection of chimpanzees from primary infection by an homologous HCV isolate has been achieved following immunization with recombinant E1 and E2 proteins (Choo et al., 1994). In that study only those chimpanzees showing high anti-E2 antibody titers were protected. While neutralizing antigenic domains were not identified, it was postulated that conformation of the immunogens was critical for the induction of neutralizing antibodies.
As there is to date no efficient in vitro replication system to grow the virus and develop neutralization assays, alternative assays to assess the biological function of anti-E1/E2 antibodies are actively searched for. Prevention of viral attachment onto presumed susceptible cells has been described in preliminary studies (Shimizu et al., 1994, Zibert et al., 1995). More recently, an “in vitro” neutralization of binding (NOB) assay has been developed that is exploiting the specific binding of a highly purified E2 protein onto susceptible target cells (Rosa et al., 1996). This assay allows the quantitative evaluation of NOB antibodies that are capable of neutralizing the binding of E2 onto such cells. Using this system, Rosa et al., have shown that only those chimpanzees immunized with E1 and E2 proteins that developed high anti-NOB titers were protected against challenged infection (Rosa et al., 1996), suggesting that NOB activity could be an indication for “in vivo” neutralization of viral infection. In HIV infection, a similar model has recently shown that affinity of antibody binding to envelope glycoprotein oligomers was a good predictor for virus neutralization (Fouts et al., 1997). Another way to assess the biological activity of anti-E1 and/or anti-E2 antibodies consists in testing the ability of such antibodies to recognize native structures believed to exist on the surface of virions. In vitro studies have shown that E1 and E2 interact to form non-covalently linked complexes (Deleersnyder et al., 1997, Ralston et al., 1993). Such complexes have been proposed to represent functional subunits of HCV virions (Deleersnyder et al., 1997, Dubuisson et al., 1994, Dubuisson and Rice, 1996, Ralston et al., 1993). Probing for the B-cell repertoire in viral infections is critical for the understanding of pathogenesis associated with these infections. Human monoclonal antibodies provide an alternative method to do so. Isolation and characterization of such antibodies have been reported in the case of HCV for only a limited number of viral antigens. These include the nucleocapsid, the NS3 and NS4 proteins (Akatsuka et al., 1993, Cerino et al., 1991, 1993, Chan et al., 1996, Mondelli et al., 1994) and more recently the glycoprotein E2 (Chan et al., 1996). In this latter case, authors used the phage display technology coupled with the use of synthetic peptides for the screening of the anti-E2 immune reactivity and were able to obtain specific IgG single-chain Fvs that recognized the E2 sequence. While a specific linear epitope sequence was identified, no biological activity for the anti-E2 antibody was described and the putative role of this antibody in the control or progression of infection remains undefined. Recently, WO97/40176 described immunoglobulin molecules obtained from a combinatorial library, which are capable of specifically binding with HCV E2 antigen. Although Fab-fragments of such immunoglobulins were demonstrated to have binding activity in a neutralization of binding assay recombinantly expressed Fab clones and corresponding whole IgG molecules were found to be negative in neutralizing the binding of the HCV E2 polypeptide.